Predicting and Understanding Consumer Financial Well-Being
Shannon Duncan
Bella Ren
Siyuan Yin
11:59 pm, May 9, 2021
1 Abstract
Financial well-being is an important part of everyday life. Past research has found that financial well-being is associated with a variety of demogrpahic, psychological, and economic factors. In this report, we explore two previously under-explored questions regarding financial well-being:
among the many economic, demographic, and psychological factors related to financial well-being, what variables are most important to include in a regression model and
can financial well-being ratings reveal meaningful clusters in long-term financial outcomes (retirement investments)?
Using LASSO and linear regression, we find that a model with just 6 variables - subjective well-being scores, financial skill scale score,s financial planning activity, confidence in ability to achieve one’s financial goals, age, and difficulty making ends meet - can account for a large amount of the variation in financial well-being. Additionally, using PCA and cluster analysis, we find that financial-well being ratings can be split into three groups that differ on their ratio of financial confidence and despair. As the ratio of confidence to despair for a group increases, the likelihood of a member of that group having retirement investments increases. Our results could help guide future researchers, policy-makers, and business people when they seek to predict or understand consumer financial well-being.
2 Introduction and Goal of Study
Financial well-being reflects one’s perceived capacity to maintain his or her “current and anticipated desired living standard and financial freedom.” (Bruggen, et al., 2017) It is a concept that is closely correlated with various domains, including financial planning, developmental psychology, and decision making (Bruggen, et al. 2017). How people feel about their financial status is a crucial topic for economists, policy makers, and other scholars to understand. Based on a report by Organization for Economic Co-operation and Development, people’s savings rate in the eurozone has dropped from 8.5% in 1999 to 6.56% in 2017 (OECD, 2016). Moreover, studies also indicate that the young generation is not prepared to manage their own finances (Avard, Manton, English, & Walker, 2005). These findings highlight the importance of understanding how people’s feelings about their financial situation change overtime.
Scholars believe that one’s financial well-being is closely associated with both more objective demographic and economic factors and subjective psychological indices. For instance, level of debt and income are strong indicators of how one feels about their current and future financial situations (Shim et al., 2009; Porter & Garman, 1992). At the same time, variations in how individuals perceive and react to their financial conditions lead people to feel differently about the same financial situations (Norvilitis, Szablicki, & Wilson, 2003). Some commonly studied psychological factors include people’s risk preference (Kim & Garman, 2003) and their belief in one’s ability to manage expenses and debt (2003). The existing literature on financial well-being suggests that the concept is a sophisticated construct that requires multifaceted perspectives to truly understand.
While these studies reveal important connections between financial well-being and other factors, many research questions remain unexplored. If financial-well being reflects a variety of both economic and psychological factors, what factors are most important to include in a model predicting it? Can financial well-being reveal meaningful clusters in the population when it comes to long-run financial health? In this project, we explore these questions using a comprehensive set of methods, including LASSO, linear regression, clustering, and principal component analysis.
3 Data Information and EDA
3.1 Data Summary
The data come from the Consumer Financial Protection Bureau’s (CFPB) National Financial Well-Being Survey, available at: https://www.consumerfinance.gov/data-research/financial-well-being-survey-data/. The survey uses CFPB’s validated financial well-being scale to measure the financial well-being of a national sample of adults selected to represent the adult population of the 50 U.S. states and the District of Columbia.
The data was collected online in English and Spanish between October 27, 2016 and December 5, 2016. In total, 6,394 participants completed the study. More specifically, 5,395 from the general population and 999 from an oversample of adults aged 62 and older.
The dataset includes 217 variables. Aside from the financial well-being scale results, it also includes measures of individual and household characteristics, income and employment, savings and safety nets, financial experiences, and behaviors, skills and attitudes that have been hypothesized to influence adults’ levels of financial well-being. For a full explanation of all variables, please see the codebook here: https://files.consumerfinance.gov/f/documents/cfpb_nfwbs-puf-codebook.pdf
3.2 Notes on Data Processing and Selection
As with any large dataset, there are a significant amount of non-substantive (i.e., “don’t recall”) or missing values. The CFPB coded these as either negative numbers or high values (98 or 99). For ease of analysis, we recoded all these values to NA.
Given the prevalence of NAs, we were not able to run LASSO using all of the potentially relevant variables in the dataset. Thus, we subset the data to a smaller set of variables. To align with the goal of the study, in doing so we focused on typical demographic and financial variables that could be related to financial well-being (age, income, ability to make ends meet, etc.), as well as psychological factors that could influence financial well-being (such as how connected one is to their future self, how materialistic they are, etc.). We only include complete observations on the variables of interest.
Additionally, for some psychological constructs of interest, the survey asked multiple questions about the same constructs. For those, we calculated a Cronbach’s alpha and if it was sufficiently high (0.6 or higher), we averaged them into one scale. We did this to make interpretation of the results cleaner and the model more parsimonious.
This left us with 2526 observations and 42 variables for our LASSO and linear regression analysis (including financial well-being).
For our PCA and clustering analysis, we used all participants who completed the financial well-being scale, so this gave us 6375 participants for that analysis.
3.3 Exploratory Data Analysis
Prior to running our main analyses, we run some exploratory data analyses in order to get a sense for our dataset. We primarily focus on the subset for LASSO and linear regression.
First, we plot a histogram of financial-well being scores. For our LASSO and linear regression, we use the IRT scores that were calculated by the CFPB according to the scale scoring criterion. We see that in our sample the scores are approximately normally distributed, ranging from 14 to 95. More details on the scale and how it was scored can be found in Appendix B.
Next, we look at income, age and retirement status. For income, we see that while the distribution is fairly flat, it is slightly skewed toward the higher end of the income scale. This may mean our sample is not as representative of the United States as the survey designers intended - given that income tends to be skewed toward the lower end. For age, the majority of participants fall between age 25 and 69, which is fairly reflective of the US population. Finally, we see that about 30% or so of the sample is retired - this is higher than the US population, but likely due to the fact that they oversampled adults 62 and older (this could also contribute to the income effects).
Then, we take a look at some of the participant’s financial habits. We see that while the majority do not have non-retirement investments, the majority have savings accounts and retirement investments.
In the subsequent charts, we look at the distributions of financial-well being scores over income and age. While we can see that there are mildly positive relationships between income and age and financial-well being, there are wide distributions of scores over all income and age intervals - indicating that other factors come into play when it comes to financial well-being. Furthermore, for age, we can see that it’s actually pretty flat until around 55 - indicating most of the gains happen later in life.
Next, we run some correlation matrices to see what the relationships between different variables look like. First, we correlate subjective assessment of one’s financial knowledge with scores on objective measures of financial knowledge. While subjective and objective knowledge are positively correlated, the correlations are small - indicating people are not great at assessing their own financial knowledge.
Finally, we correlate financial-well being scores with different demographic, economic and psychological variables. We see that while financial well-being is correlated with most of these variables, the correlations range in strength and direction. No relationship has a perfect 1:1 relationship with financial well-being. As one might expect, it’s likely that financial-well being is influenced by many different factors - not just income or other financial habits. We will now move on to our formal data analysis to get a sense for what factors best predict financial well-being.
4 Data Analysis
Now that we have gotten a better sense for our dataset, we move on to our formal data analyses. We start with LASSO and linear regression, hoping to answer the question: what factors are most important to include in a model predicting financial well-being scores?
4.1 LASSO and Linear Regression
4.1.1 LASSO
We use LASSO to help narrow down the predictors and create the most parsimonious and interpretable model. LASSO (Least Absolute Shrinkage and Selection Operator) is a model selection scheme that used to find a sparse model. It produces a restricted least squared solution by adding constraints over the parameters.
## (Intercept) SWBscore FSscore ActScore
## 24.203174902 1.456971867 0.205151581 2.020138125
## FINGOALS PropPlanscore ManageScore FRUGALITY
## -0.665607468 -1.285546332 1.535737732 -0.115647121
## ASK1_1 GOALCONF KHscore agecat
## -0.617431373 2.759799770 0.164051634 0.769421340
## PPHHSIZE PPINCIMP EMPLOY1_8 ENDSMEET
## -0.102496424 0.558043451 2.523322844 -7.706179392
## SHOCKS_12 VOLATILITY MaterialismScore CONNECT
## 0.521063771 -0.212202694 -0.077131690 0.005372795
## SCFHORIZON DISCOUNT
## 0.563424724 0.388841244LASSO indicates that we should include at minimum 21 variables in our model. These variables include (in order): subjective well-being scores, financial skill scale scores, financial follow through scores, presence of a financial goal, financial planning activity, financial management, frugality, doing research before making financial decisions, confidence in ability to achieve one’s financial goals, Knoll and Houts financial knowledge scale scores, age, households size, household income, retirement status, difficulty making ends meet, presence of financial shocks, household income volatility, materialism, psychological connectedness, financial planning time horizon, and discount rate.
| Lasso Model | |
|---|---|
| (Intercept) | 26.1852*** |
| (2.2379) | |
| SWBscore | 1.6829*** |
| (0.1557) | |
| FSscore | 0.2554*** |
| (0.0197) | |
| ActScore | 2.2919*** |
| (0.3506) | |
| FINGOALS | -1.1472** |
| (0.3673) | |
| PropPlanscore | -1.8657*** |
| (0.2333) | |
| ManageScore | 1.6353*** |
| (0.2945) | |
| FRUGALITY | -0.8033*** |
| (0.2096) | |
| ASK1_1 | -1.3047*** |
| (0.1866) | |
| GOALCONF | 2.9109*** |
| (0.3173) | |
| KHscore | 0.8038· |
| (0.4148) | |
| agecat | 0.7467*** |
| (0.1168) | |
| PPHHSIZE | -0.4021** |
| (0.1477) | |
| PPINCIMP | 0.6785*** |
| (0.0787) | |
| EMPLOY1_8 | 2.8772*** |
| (0.4833) | |
| ENDSMEET | -7.1948*** |
| (0.3739) | |
| SHOCKS_12 | 0.8857** |
| (0.3158) | |
| VOLATILITY | -0.8147** |
| (0.2915) | |
| MaterialismScore | -0.6474*** |
| (0.1955) | |
| CONNECT | 0.0102 |
| (0.0067) | |
| SCFHORIZON | 0.6682*** |
| (0.1344) | |
| DISCOUNT | 0.9538** |
| (0.3593) | |
| R2 | 0.6990 |
| Adj. R2 | 0.6964 |
| Num. obs. | 2526 |
| p < 0.001; p < 0.01; p < 0.05; ·p < 0.1 | |
4.1.2 Fine-Tuning the Lasso Model
While LASSO produces a relatively sparse model, we can see that this model still has quite a few variables, and not all variables are statistically significant at the 0.05 level (psychological connectedness). Thus, we further fine-tune our with model selection using Mallow’s Cp. Our goal is to find the model with the smallest value of Cp. We use the elbow rule, or select the model at the point where the scree plot begins to level off.
[1] "SWBscore" "FSscore" "PropPlanscore" "GOALCONF"
[5] "agecat" "ENDSMEET"The screeplot begins to level off around 6 predictors, so we chose to use a model with 6 predictors as our final model. These 6 predictors, as listed above, are: subjective well-being scores, financial skill scale scores, financial planning activity, confidence in ability to achieve one’s financial goals, age, and difficulty making ends meet.
4.1.3 Final Model
| Final Model | |
|---|---|
| (Intercept) | 32.6949*** |
| (1.5853) | |
| SWBscore | 1.9565*** |
| (0.1642) | |
| FSscore | 0.3095*** |
| (0.0184) | |
| PropPlanscore | -2.2693*** |
| (0.2255) | |
| GOALCONF | 4.0320*** |
| (0.3285) | |
| agecat | 1.6028*** |
| (0.0814) | |
| ENDSMEET | -10.0146*** |
| (0.3625) | |
| R2 | 0.6495 |
| Adj. R2 | 0.6486 |
| Num. obs. | 2526 |
| p < 0.001; p < 0.01; p < 0.05; ·p < 0.1 | |
We can see that our final model is much more parsimonious than the original LASSO model with 21 variables, but the R^2 is not too different - 0.6495 for our 6 variable model and 0.6990 for our 21 variable model.
The model reasonably meets the assumptions of the linear model (see Appendix C for more detail).
Our final model shows that one’s subjective well-being, financial skills, confidence in ability to reach one’s goals, and age positively predict financial well-being. In other words, as these variables increase, financial well-being increases. On the other hand, financial planning activity and ability to make ends meet have a negative relationship with financial well-being - as planning and difficulty making ends meet increase, financial well-being decreases.
4.2 PCA And Cluster Analysis
Now that we know what the most parsimonious model predicting financial well-being scores is, we turn to understanding how financial well-being can reveal interesting groupings in the population on long-term financial outcomes. More specifically, to answer this question, we conduct PCA and Cluster Analysis on the financial well-being scale and relate it the presence of retirement investments. Conducting PCA and Cluster Analysis makes it so that we can no longer use the scores generated by the scale scoring criterion. Thus, we turn to using the raw answers to each of the scale questions. The scale consists of 10 questions, which can be found in Appendix B. As note above, for this analysis, we use all participants who completed all 10 scale questions.
4.2.1 PCA
Prior to clustering, we conduct PCA on the 10 scale questions. While we obtain 10 PCs from the data, here we focus on we first two, which by definition capture the greatest amount of variation on the data (compared to the other PCs).
We see that PC2 looks like a weighted average of all the questions, with some questions’ answers being weighted more heavily than others. PC1, however, is a little more complex. As shown in the results, PC1 has loadings that give positive weight to FWB1_1, FWB1_2, FWB1_4, FWB2_2. The questions related to these items are all positively related to financial confidence (e.g. “I could handle a major unexpected expense”). Meanwhile, all the other items are more representative of financial despair or lack of confidence financially (e.g. “I am behind with my finances”). Thus, it seems that those with higher PC1 scores have more confidence than they do despair when it comes to financial behaviors, and those with lower scores have more financial despair than they have confidence.
| PC1 | PC2 | |
|---|---|---|
| FWB1_1 | 0.3523152 | -0.3936878 |
| FWB1_2 | 0.2986641 | -0.3767554 |
| FWB1_3 | -0.3296087 | -0.3097390 |
| FWB1_4 | 0.2838223 | -0.3337840 |
| FWB1_5 | -0.3001864 | -0.5308580 |
| FWB1_6 | -0.2974687 | -0.2986642 |
| FWB2_1 | -0.3453675 | -0.0629046 |
| FWB2_2 | 0.3621352 | -0.2737495 |
| FWB2_3 | -0.2876497 | -0.0400610 |
| FWB2_4 | -0.2927699 | -0.2102478 |
To get a sense for how the PCs generally relate to retirement investments, we do some initial EDA with the PC scores. Since the CFPB uses an IRT score for financial well being scores, and its calculation is different based on one’s age group, we also include this in our brief EDA. Overall, it seems people who have a retirement account have higher financial well-being on PC1 on average. However, for PC2, there do not seem to be large differences in holding retirement investments and PC2 score distributions. The figure suggest that these relationships are pretty consistent across all age groups.
4.2.2 Cluster Analysis
Now that we have a sense for the PCs, we move on to cluster analysis. We start by calculating the total within-clusters sum of squares for different numbers of clusters to determine how many clusters we should use to group the data. We then plot this on a scree plot and use the elbow rule to select our number of clusters. Based on the graph, we use 3 clusters.
In figure 1, we explore how the three clusters differ in their PC1 and PC2 scores. The figure shows that all three clusters have similar PC2 scores. In other words, the weighted averages on the financial well-being scale are similar across clusters. However, on average, group 2 has the lowest PC1 scores, group 1 has slightly higher PC1 scores, and group 3 has the highest PC1 scores. If we refer back to our PC1 interpretation above, we can understand how these clusters vary. Given that higher PC1 scores suggest higher financial confidence while lower PC1 scores suggest higher financial despair, people who have the highest PC1 scores (people in group 3) have higher financial confidence than despair, followed by group 1 (who fall somewhere in the middle - about evenly confident/despairing), then by group 2 (who have more financial despair than confidence).
Next, we investigate how these 3 different financial well-being clusters differ on their retirement saving status. The figures illustrate that most people in group 2 do not have a retirement savings account; relatively more people in group 1 have a retirement savings account (more of an even split); lastly, most people in group 3 have a retirement saving account. Relating this finding to the prior figures, we find that people who are more confident about their financial well beings are more likely to have a retirement savings account. We can see that there is some association with age such that group 2 has more younger people, while groups 1 and 3 are pretty evenly distributed age-wise, comparatively. Thus, it seems that the clusters based on confidence/despair are more meaningful than age alone when it comes to the retirement savings pattern.
5 Conclusion
In this report, we sought to answer two questions. First, among the many economic and psychological factors related to financial well-being, what variables are most important to include in a regression model? Second, can financial well-being ratings reveal meaningful clusters in long-term financial outcomes (retirement investments)?
Using LASSO and linear regression methods, we find that the most parsimonious model of consumer financial well-being includes 6 variables: subjective well-being scores, financial skill scale scores, financial planning activity, confidence in ability to achieve one’s financial goals, age, and difficulty making ends meet. Our final model shows that as one’s subjective well-being, financial skills, confidence in ability to reach one’s goals, and age increase, financial well-being increases. On the other hand, as financial planning activity and ability to make ends meet increase, financial well-being decreases. Of these results, the negative association between financial planning activity and financial well-being is most interesting. One might expect that planning would improve financial well-being - allowing for greater savings, discretionary spending, etc. However, it’s possible that planning more may actually make one feel more stressed and less confident about their finances. Seeing the numbers all the time may make us feel like we are worse off than we are. Or simple feeling like we need to plan may make us feel worse off if we believe that richer people can just live and let be. Future research can explore this idea.
Using PCA and Cluster Analysis, we find that financial well-being scores can be split into three clusters. These three clusters vary in regards to their financial confidence and despair. Clusters with a higher ratio of confidence relative to despair have better long-term financial outcomes, meaning that they are more likely to have retirement investments. These results and interpretations also align with our regression analysis, in which confidence in one’s financial ability was an important predictor to include in modeling financial well-being. Future research could explore other meaningful relationships between these clusters and other outcomes.
Overall, our results have a few important implications. First, they support that both economic and psychological factors are important for financial well-being and long-term financial health. Second, given that confidence is important in both our linear regression and our clustering analysis, they suggest that confidence is an important factor to prioritize when considering consumer financial well-being and health. Finally, they show that despite the vast literature detailing many different variables associated with financial well-being, a model with just 6 variables can do a good job accounting for a large amount of the variation in well-being. As a whole, these results could help guide future researchers, policy-makers, and business people when they seek to predict or understand consumer financial well-being.
6 References
Avard, S., Manton, E., English, D., & Walker, J. (2005). The financial knowledge of college freshmen. College student journal, 39(2), 321-340.
Brüggen, E. C., Hogreve, J., Holmlund, M., Kabadayi, S., & Löfgren, M. (2017). Financial well-being: A conceptualization and research agenda. Journal of Business Research, 79, 228-237.
Kim, J., & Garman, E. T. (2003). Financial stress and absenteeism: An empirically derived model. Financial Counseling and Planning, 14(1), 31–42.
Norvilitis, J. M., Szablicki, P. B., & Wilson, S. D. (2003). Factors influencing levels of credit- card debt in college students. Journal of Applied Social Psychology, 33(5), 935–947.
OECD (2016). Household savings forecast (Indicator). retrieved from https://data.oecd. org/hha/household-savings-forecast.html
Porter, N. M., & Garman, E. T. (1992). Money as part of a measure of financial well-being. American Behavioral Scientist, 35(6), 820–826.
Shim, S., Xiao, J. J., Barber, B. L., & Lyons, A. C. (2009). Pathways to life success: A concep- tual model of financial well-being for young adults. Journal of Applied Developmental Psychology, 30(6), 708–723.
7 Appendix
7.1 Appendix A: Summary of Entire Dataset Before Cleaning
| Name | cfpbdatanorecode |
| Number of rows | 6394 |
| Number of columns | 217 |
| _______________________ | |
| Column type frequency: | |
| numeric | 217 |
| ________________________ | |
| Group variables | None |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| PUF_ID | 0 | 1 | 10892.39 | 1967.85 | 7123.00 | 9235.25 | 10901.50 | 12570.75 | 14400.00 | ▆▇▇▇▇ |
| sample | 0 | 1 | 1.28 | 0.57 | 1.00 | 1.00 | 1.00 | 1.00 | 3.00 | ▇▁▂▁▁ |
| fpl | 0 | 1 | 2.66 | 0.66 | 1.00 | 3.00 | 3.00 | 3.00 | 3.00 | ▁▁▂▁▇ |
| SWB_1 | 0 | 1 | 5.35 | 1.50 | -4.00 | 5.00 | 6.00 | 6.00 | 7.00 | ▁▁▁▂▇ |
| SWB_2 | 0 | 1 | 5.36 | 1.54 | -4.00 | 5.00 | 6.00 | 7.00 | 7.00 | ▁▁▁▂▇ |
| SWB_3 | 0 | 1 | 5.43 | 1.61 | -4.00 | 5.00 | 6.00 | 7.00 | 7.00 | ▁▁▁▂▇ |
| FWBscore | 0 | 1 | 56.03 | 14.15 | -4.00 | 48.00 | 56.00 | 65.00 | 95.00 | ▁▁▇▇▂ |
| FWB1_1 | 0 | 1 | 3.05 | 1.24 | -4.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▁▁▂▇▆ |
| FWB1_2 | 0 | 1 | 3.19 | 1.11 | -4.00 | 3.00 | 3.00 | 4.00 | 5.00 | ▁▁▁▇▆ |
| FWB1_3 | 0 | 1 | 2.53 | 1.20 | -4.00 | 2.00 | 2.00 | 3.00 | 5.00 | ▁▁▃▇▂ |
| FWB1_4 | 0 | 1 | 3.29 | 1.05 | -4.00 | 3.00 | 3.00 | 4.00 | 5.00 | ▁▁▁▇▆ |
| FWB1_5 | 0 | 1 | 2.77 | 1.27 | -4.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▁▁▃▇▅ |
| FWB1_6 | 0 | 1 | 3.07 | 1.18 | -4.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▁▁▁▇▅ |
| FWB2_1 | 0 | 1 | 2.34 | 1.18 | -4.00 | 1.00 | 2.00 | 3.00 | 5.00 | ▁▁▃▇▂ |
| FWB2_2 | 0 | 1 | 3.39 | 1.27 | -4.00 | 3.00 | 3.00 | 4.00 | 5.00 | ▁▁▂▇▇ |
| FWB2_3 | 0 | 1 | 2.03 | 1.11 | -4.00 | 1.00 | 2.00 | 3.00 | 5.00 | ▁▁▆▇▂ |
| FWB2_4 | 0 | 1 | 2.69 | 1.14 | -4.00 | 2.00 | 3.00 | 3.00 | 5.00 | ▁▁▂▇▃ |
| FSscore | 0 | 1 | 50.72 | 12.66 | -1.00 | 42.00 | 50.00 | 57.00 | 85.00 | ▁▁▇▇▂ |
| FS1_1 | 0 | 1 | 3.61 | 0.95 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▅▇ |
| FS1_2 | 0 | 1 | 3.54 | 1.06 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▂▅▇ |
| FS1_3 | 0 | 1 | 3.24 | 1.03 | -1.00 | 3.00 | 3.00 | 4.00 | 5.00 | ▁▁▃▇▇ |
| FS1_4 | 0 | 1 | 3.32 | 0.96 | -1.00 | 3.00 | 3.00 | 4.00 | 5.00 | ▁▁▂▇▇ |
| FS1_5 | 0 | 1 | 3.05 | 1.02 | -1.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▁▂▃▇▆ |
| FS1_6 | 0 | 1 | 3.76 | 0.95 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▃▇ |
| FS1_7 | 0 | 1 | 3.70 | 0.99 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▃▇ |
| FS2_1 | 0 | 1 | 3.74 | 0.89 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▃▇ |
| FS2_2 | 0 | 1 | 3.63 | 0.96 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▅▇ |
| FS2_3 | 0 | 1 | 2.67 | 0.93 | -1.00 | 2.00 | 3.00 | 3.00 | 5.00 | ▁▂▆▇▃ |
| SUBKNOWL1 | 0 | 1 | 4.67 | 1.28 | -1.00 | 4.00 | 5.00 | 5.00 | 7.00 | ▁▁▁▇▃ |
| ACT1_1 | 0 | 1 | 4.21 | 0.90 | -1.00 | 4.00 | 4.00 | 5.00 | 5.00 | ▁▁▁▂▇ |
| ACT1_2 | 0 | 1 | 3.61 | 0.93 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▅▇ |
| FINGOALS | 0 | 1 | 0.62 | 0.51 | -1.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▁▁▅▁▇ |
| PROPPLAN_1 | 0 | 1 | 3.67 | 1.04 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▂▂▇ |
| PROPPLAN_2 | 0 | 1 | 3.63 | 0.95 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▃▇ |
| PROPPLAN_3 | 0 | 1 | 3.67 | 0.92 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▃▇ |
| PROPPLAN_4 | 0 | 1 | 3.25 | 1.03 | -1.00 | 3.00 | 3.00 | 4.00 | 5.00 | ▁▁▃▆▇ |
| MANAGE1_1 | 0 | 1 | 4.53 | 0.92 | -1.00 | 4.00 | 5.00 | 5.00 | 5.00 | ▁▁▁▁▇ |
| MANAGE1_2 | 0 | 1 | 3.81 | 1.08 | -1.00 | 3.00 | 4.00 | 5.00 | 5.00 | ▁▁▁▂▇ |
| MANAGE1_3 | 0 | 1 | 3.54 | 1.58 | -1.00 | 2.00 | 4.00 | 5.00 | 5.00 | ▁▂▂▂▇ |
| MANAGE1_4 | 0 | 1 | 4.20 | 1.07 | -1.00 | 4.00 | 5.00 | 5.00 | 5.00 | ▁▁▁▁▇ |
| SAVEHABIT | 0 | 1 | 4.37 | 1.49 | -1.00 | 4.00 | 5.00 | 6.00 | 6.00 | ▁▁▃▃▇ |
| FRUGALITY | 0 | 1 | 5.21 | 0.93 | -1.00 | 5.00 | 5.00 | 6.00 | 6.00 | ▁▁▁▁▇ |
| AUTOMATED_1 | 0 | 1 | 2.30 | 2.91 | -1.00 | 0.00 | 1.00 | 7.00 | 7.00 | ▆▇▁▁▆ |
| AUTOMATED_2 | 0 | 1 | 1.95 | 2.75 | -1.00 | 0.00 | 1.00 | 1.00 | 7.00 | ▇▇▁▁▅ |
| ASK1_1 | 0 | 1 | 3.70 | 1.06 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▃▇ |
| ASK1_2 | 0 | 1 | 2.83 | 1.05 | -1.00 | 2.00 | 3.00 | 3.00 | 5.00 | ▁▂▅▇▅ |
| SUBNUMERACY2 | 0 | 1 | 3.67 | 1.68 | -1.00 | 3.00 | 4.00 | 5.00 | 6.00 | ▁▃▆▅▇ |
| SUBNUMERACY1 | 0 | 1 | 4.33 | 1.45 | -1.00 | 4.00 | 5.00 | 5.00 | 6.00 | ▁▁▃▃▇ |
| CHANGEABLE | 0 | 1 | 3.74 | 1.62 | -1.00 | 2.00 | 4.00 | 5.00 | 7.00 | ▁▅▃▇▃ |
| GOALCONF | 0 | 1 | 3.21 | 0.78 | -1.00 | 3.00 | 3.00 | 4.00 | 4.00 | ▁▁▂▇▆ |
| LMscore | 0 | 1 | 2.51 | 0.76 | 0.00 | 2.00 | 3.00 | 3.00 | 3.00 | ▁▁▁▃▇ |
| FINKNOWL1 | 0 | 1 | 1.18 | 0.54 | -1.00 | 1.00 | 1.00 | 1.00 | 3.00 | ▁▁▇▁▁ |
| FINKNOWL2 | 0 | 1 | 2.66 | 0.69 | -1.00 | 3.00 | 3.00 | 3.00 | 3.00 | ▁▁▁▁▇ |
| FINKNOWL3 | 0 | 1 | 1.86 | 0.41 | -1.00 | 2.00 | 2.00 | 2.00 | 2.00 | ▁▁▁▁▇ |
| FK1correct | 0 | 1 | 0.86 | 0.34 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▁▁▁▁▇ |
| FK2correct | 0 | 1 | 0.77 | 0.42 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▂▁▁▁▇ |
| FK3correct | 0 | 1 | 0.87 | 0.34 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▁▁▁▁▇ |
| KHscore | 0 | 1 | -0.06 | 0.81 | -2.05 | -0.57 | -0.19 | 0.71 | 1.27 | ▂▅▇▆▇ |
| KHKNOWL1 | 0 | 1 | 2.50 | 0.71 | -1.00 | 2.00 | 3.00 | 3.00 | 3.00 | ▁▁▁▃▇ |
| KHKNOWL2 | 0 | 1 | 2.76 | 0.62 | -1.00 | 3.00 | 3.00 | 3.00 | 3.00 | ▁▁▁▁▇ |
| KHKNOWL3 | 0 | 1 | 1.96 | 0.64 | -1.00 | 2.00 | 2.00 | 2.00 | 3.00 | ▁▁▂▇▂ |
| KHKNOWL4 | 0 | 1 | 1.15 | 0.41 | -1.00 | 1.00 | 1.00 | 1.00 | 2.00 | ▁▁▁▇▂ |
| KHKNOWL5 | 0 | 1 | 1.23 | 0.47 | -1.00 | 1.00 | 1.00 | 1.00 | 2.00 | ▁▁▁▇▂ |
| KHKNOWL6 | 0 | 1 | 1.91 | 0.33 | -1.00 | 2.00 | 2.00 | 2.00 | 2.00 | ▁▁▁▁▇ |
| KHKNOWL7 | 0 | 1 | 3.12 | 0.95 | -1.00 | 2.00 | 3.00 | 4.00 | 4.00 | ▁▁▃▆▇ |
| KHKNOWL8 | 0 | 1 | 2.29 | 1.15 | -1.00 | 1.00 | 2.00 | 3.00 | 4.00 | ▁▆▇▃▆ |
| KHKNOWL9 | 0 | 1 | 1.08 | 0.33 | -1.00 | 1.00 | 1.00 | 1.00 | 2.00 | ▁▁▁▇▁ |
| KH1correct | 0 | 1 | 0.60 | 0.49 | 0.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▅▁▁▁▇ |
| KH2correct | 0 | 1 | 0.84 | 0.36 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▂▁▁▁▇ |
| KH3correct | 0 | 1 | 0.67 | 0.47 | 0.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▃▁▁▁▇ |
| KH4correct | 0 | 1 | 0.83 | 0.38 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▂▁▁▁▇ |
| KH5correct | 0 | 1 | 0.75 | 0.43 | 0.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▃▁▁▁▇ |
| KH6correct | 0 | 1 | 0.92 | 0.27 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▁▁▁▁▇ |
| KH7correct | 0 | 1 | 0.44 | 0.50 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▆ |
| KH8correct | 0 | 1 | 0.35 | 0.48 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▅ |
| KH9correct | 0 | 1 | 0.90 | 0.30 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▁▁▁▁▇ |
| ENDSMEET | 0 | 1 | 1.44 | 0.66 | -1.00 | 1.00 | 1.00 | 2.00 | 3.00 | ▁▁▇▅▁ |
| HOUSING | 0 | 1 | 1.41 | 0.67 | -1.00 | 1.00 | 1.00 | 2.00 | 3.00 | ▁▁▇▃▁ |
| LIVINGARRANGEMENT | 0 | 1 | 2.08 | 0.86 | -1.00 | 2.00 | 2.00 | 2.00 | 5.00 | ▁▂▇▁▁ |
| HOUSERANGES | 0 | 1 | 16.66 | 32.64 | -1.00 | 2.00 | 4.00 | 6.00 | 99.00 | ▇▁▁▁▁ |
| IMPUTATION_FLAG | 0 | 1 | 0.06 | 0.24 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| VALUERANGES | 0 | 1 | 4.98 | 19.93 | -2.00 | -2.00 | 1.00 | 3.00 | 99.00 | ▇▁▁▁▁ |
| MORTGAGE | 0 | 1 | 9.72 | 29.07 | -2.00 | -2.00 | 1.00 | 2.00 | 99.00 | ▇▁▁▁▁ |
| SAVINGSRANGES | 0 | 1 | 21.37 | 36.26 | -1.00 | 4.00 | 5.00 | 7.00 | 99.00 | ▇▁▁▁▂ |
| PRODHAVE_1 | 0 | 1 | 0.85 | 0.35 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▂▁▁▁▇ |
| PRODHAVE_2 | 0 | 1 | 0.52 | 0.50 | 0.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▇▁▁▁▇ |
| PRODHAVE_3 | 0 | 1 | 0.71 | 0.45 | 0.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▃▁▁▁▇ |
| PRODHAVE_4 | 0 | 1 | 0.58 | 0.49 | 0.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▆▁▁▁▇ |
| PRODHAVE_5 | 0 | 1 | 0.34 | 0.47 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▅ |
| PRODHAVE_6 | 0 | 1 | 0.31 | 0.46 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▃ |
| PRODHAVE_7 | 0 | 1 | 0.06 | 0.24 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PRODHAVE_8 | 0 | 1 | 0.14 | 0.35 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PRODHAVE_9 | 0 | 1 | 0.04 | 0.20 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PRODUSE_1 | 0 | 1 | 0.03 | 0.16 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PRODUSE_2 | 0 | 1 | 0.02 | 0.13 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PRODUSE_3 | 0 | 1 | 0.09 | 0.28 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PRODUSE_4 | 0 | 1 | 0.04 | 0.20 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PRODUSE_5 | 0 | 1 | 0.07 | 0.25 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PRODUSE_6 | 0 | 1 | 0.80 | 0.40 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▂▁▁▁▇ |
| CONSPROTECT1 | 0 | 1 | 1.79 | 0.83 | -1.00 | 1.00 | 2.00 | 2.00 | 4.00 | ▁▇▇▃▁ |
| CONSPROTECT2 | 0 | 1 | 1.50 | 0.67 | -1.00 | 1.00 | 1.00 | 2.00 | 3.00 | ▁▁▇▆▁ |
| CONSPROTECT3 | 0 | 1 | 0.06 | 0.27 | -1.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▁▁▇▁▁ |
| EARNERS | 0 | 1 | 1.63 | 0.64 | -1.00 | 1.00 | 2.00 | 2.00 | 3.00 | ▁▁▆▇▁ |
| VOLATILITY | 0 | 1 | 1.31 | 0.64 | -1.00 | 1.00 | 1.00 | 2.00 | 3.00 | ▁▁▇▂▁ |
| SNAP | 0 | 1 | 0.23 | 1.09 | -1.00 | 0.00 | 0.00 | 0.00 | 8.00 | ▇▁▁▁▁ |
| MATHARDSHIP_1 | 0 | 1 | 1.22 | 0.53 | -1.00 | 1.00 | 1.00 | 1.00 | 3.00 | ▁▁▇▁▁ |
| MATHARDSHIP_2 | 0 | 1 | 1.19 | 0.50 | -1.00 | 1.00 | 1.00 | 1.00 | 3.00 | ▁▁▇▁▁ |
| MATHARDSHIP_3 | 0 | 1 | 1.11 | 0.42 | -1.00 | 1.00 | 1.00 | 1.00 | 3.00 | ▁▁▇▁▁ |
| MATHARDSHIP_4 | 0 | 1 | 1.19 | 0.50 | -1.00 | 1.00 | 1.00 | 1.00 | 3.00 | ▁▁▇▁▁ |
| MATHARDSHIP_5 | 0 | 1 | 1.17 | 0.47 | -1.00 | 1.00 | 1.00 | 1.00 | 3.00 | ▁▁▇▁▁ |
| MATHARDSHIP_6 | 0 | 1 | 1.07 | 0.34 | -1.00 | 1.00 | 1.00 | 1.00 | 3.00 | ▁▁▇▁▁ |
| COLLECT | 0 | 1 | 0.42 | 1.53 | -1.00 | 0.00 | 0.00 | 0.00 | 8.00 | ▇▁▁▁▁ |
| REJECTED_1 | 0 | 1 | 0.09 | 0.32 | -1.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▁▁▇▁▁ |
| REJECTED_2 | 0 | 1 | 0.10 | 0.36 | -1.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▁▁▇▁▁ |
| ABSORBSHOCK | 0 | 1 | 3.52 | 1.51 | -1.00 | 3.00 | 4.00 | 4.00 | 8.00 | ▁▂▇▁▁ |
| BENEFITS_1 | 0 | 1 | 0.70 | 0.47 | -1.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▁▁▃▁▇ |
| BENEFITS_2 | 0 | 1 | 0.53 | 0.51 | -1.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▁▁▇▁▇ |
| BENEFITS_3 | 0 | 1 | 0.32 | 0.48 | -1.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▁▁▇▁▃ |
| BENEFITS_4 | 0 | 1 | 0.16 | 0.38 | -1.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▁▁▇▁▂ |
| BENEFITS_5 | 0 | 1 | 0.43 | 0.51 | -1.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▁▁▇▁▆ |
| FRAUD2 | 0 | 1 | 0.90 | 2.14 | -1.00 | 0.00 | 0.00 | 1.00 | 8.00 | ▇▃▁▁▁ |
| COVERCOSTS | 0 | 1 | 1.97 | 0.78 | -1.00 | 2.00 | 2.00 | 2.00 | 4.00 | ▁▃▇▂▁ |
| BORROW_1 | 0 | 1 | 0.56 | 0.55 | -1.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▁▁▅▁▇ |
| BORROW_2 | 0 | 1 | 0.27 | 0.53 | -1.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▁▁▇▁▃ |
| SHOCKS_1 | 0 | 1 | 0.06 | 0.24 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_2 | 0 | 1 | 0.07 | 0.25 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_3 | 0 | 1 | 0.01 | 0.09 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_4 | 0 | 1 | 0.21 | 0.40 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▂ |
| SHOCKS_5 | 0 | 1 | 0.14 | 0.35 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_6 | 0 | 1 | 0.02 | 0.14 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_7 | 0 | 1 | 0.03 | 0.18 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_8 | 0 | 1 | 0.01 | 0.10 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_9 | 0 | 1 | 0.06 | 0.25 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_10 | 0 | 1 | 0.04 | 0.19 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_11 | 0 | 1 | 0.13 | 0.34 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| SHOCKS_12 | 0 | 1 | 0.50 | 0.50 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▇ |
| MANAGE2 | 0 | 1 | 2.35 | 0.78 | -1.00 | 2.00 | 3.00 | 3.00 | 3.00 | ▁▁▂▆▇ |
| PAIDHELP | 0 | 1 | -1.00 | 1.05 | -2.00 | -2.00 | -2.00 | 0.00 | 1.00 | ▇▁▁▇▁ |
| HSLOC | 0 | 1 | 0.87 | 0.78 | -1.00 | 1.00 | 1.00 | 1.00 | 8.00 | ▁▇▁▁▁ |
| PAREDUC | 0 | 1 | 2.92 | 1.28 | -1.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▁▂▇▆▇ |
| FINSOC2_1 | 0 | 1 | 0.34 | 0.48 | -1.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▁▁▇▁▅ |
| FINSOC2_2 | 0 | 1 | 0.65 | 0.48 | -1.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▁▁▅▁▇ |
| FINSOC2_3 | 0 | 1 | 0.36 | 0.49 | -1.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▁▁▇▁▅ |
| FINSOC2_4 | 0 | 1 | 0.60 | 0.50 | -1.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▁▁▅▁▇ |
| FINSOC2_5 | 0 | 1 | 0.74 | 0.45 | -1.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▁▁▃▁▇ |
| FINSOC2_6 | 0 | 1 | 0.40 | 0.50 | -1.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▁▁▇▁▆ |
| FINSOC2_7 | 0 | 1 | 0.42 | 0.50 | -1.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▁▁▇▁▆ |
| OBJNUMERACY1 | 0 | 1 | 1.92 | 0.71 | -1.00 | 2.00 | 2.00 | 2.00 | 3.00 | ▁▁▁▇▁ |
| ON2correct | 0 | 1 | 0.66 | 0.47 | 0.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▅▁▁▁▇ |
| ON1correct | 0 | 1 | 0.78 | 0.42 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▂▁▁▁▇ |
| MATERIALISM_1 | 0 | 1 | 2.54 | 1.19 | -1.00 | 2.00 | 3.00 | 3.00 | 5.00 | ▁▃▅▇▃ |
| MATERIALISM_2 | 0 | 1 | 2.89 | 1.18 | -1.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▁▂▅▇▆ |
| MATERIALISM_3 | 0 | 1 | 2.29 | 1.15 | -1.00 | 2.00 | 2.00 | 3.00 | 5.00 | ▁▆▇▇▃ |
| CONNECT | 0 | 1 | 68.15 | 34.07 | -1.00 | 50.00 | 80.00 | 97.00 | 100.00 | ▂▁▂▂▇ |
| HEALTH | 0 | 1 | 3.41 | 1.01 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▂▅▇ |
| SCFHORIZON | 0 | 1 | 3.06 | 1.38 | -1.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▁▃▃▅▇ |
| DISCOUNT | 0 | 1 | 1.57 | 0.56 | -1.00 | 1.00 | 2.00 | 2.00 | 2.00 | ▁▁▁▅▇ |
| MEMLOSS | 0 | 1 | 0.09 | 0.33 | -1.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▁▁▇▁▁ |
| DISTRESS | 0 | 1 | 3.11 | 1.14 | -1.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▁▁▅▆▇ |
| SELFCONTROL_1 | 0 | 1 | 1.91 | 0.85 | -1.00 | 1.00 | 2.00 | 2.00 | 4.00 | ▁▆▇▃▁ |
| SELFCONTROL_2 | 0 | 1 | 2.86 | 0.77 | -1.00 | 2.00 | 3.00 | 3.00 | 4.00 | ▁▁▃▇▂ |
| SELFCONTROL_3 | 0 | 1 | 3.02 | 0.76 | -1.00 | 3.00 | 3.00 | 3.00 | 4.00 | ▁▁▂▇▃ |
| OUTLOOK_1 | 0 | 1 | 3.51 | 1.18 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▁▅▇ |
| OUTLOOK_2 | 0 | 1 | 3.37 | 1.27 | -1.00 | 3.00 | 4.00 | 4.00 | 5.00 | ▁▁▂▅▇ |
| INTERCONNECTIONS_1 | 0 | 1 | 0.22 | 0.41 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▂ |
| INTERCONNECTIONS_2 | 0 | 1 | 0.45 | 0.50 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▆ |
| INTERCONNECTIONS_3 | 0 | 1 | 0.14 | 0.35 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| INTERCONNECTIONS_4 | 0 | 1 | 0.06 | 0.23 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| INTERCONNECTIONS_5 | 0 | 1 | 0.23 | 0.42 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▂ |
| INTERCONNECTIONS_6 | 0 | 1 | 0.04 | 0.19 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| INTERCONNECTIONS_7 | 0 | 1 | 0.20 | 0.40 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▂ |
| INTERCONNECTIONS_8 | 0 | 1 | 0.26 | 0.44 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▃ |
| INTERCONNECTIONS_9 | 0 | 1 | 0.03 | 0.18 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| INTERCONNECTIONS_10 | 0 | 1 | 0.19 | 0.39 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▂ |
| PEM | 0 | 1 | 4.64 | 1.75 | -1.00 | 3.00 | 5.00 | 6.00 | 7.00 | ▁▂▃▇▇ |
| HOUSESAT | 0 | 1 | 3.30 | 0.86 | -1.00 | 3.00 | 3.00 | 4.00 | 4.00 | ▁▁▂▆▇ |
| SOCSEC1 | 0 | 1 | -1.01 | 1.37 | -2.00 | -2.00 | -2.00 | 1.00 | 1.00 | ▇▁▁▁▃ |
| SOCSEC2 | 0 | 1 | 15.36 | 29.06 | -3.00 | -2.00 | -2.00 | 62.00 | 70.00 | ▇▁▁▁▃ |
| SOCSEC3 | 0 | 1 | 39.94 | 33.25 | -3.00 | -2.00 | 62.00 | 67.00 | 71.00 | ▅▁▁▁▇ |
| LIFEEXPECT | 0 | 1 | 64.41 | 36.61 | -2.00 | 50.00 | 80.00 | 99.00 | 100.00 | ▃▁▂▂▇ |
| HHEDUC | 0 | 1 | 3.50 | 1.22 | -1.00 | 3.00 | 4.00 | 5.00 | 5.00 | ▁▁▃▅▇ |
| KIDS_NoChildren | 0 | 1 | 0.48 | 0.66 | -1.00 | 0.00 | 1.00 | 1.00 | 1.00 | ▁▁▅▁▇ |
| KIDS_1 | 0 | 1 | 0.16 | 0.47 | -1.00 | 0.00 | 0.00 | 0.00 | 2.00 | ▁▇▁▁▁ |
| KIDS_2 | 0 | 1 | 0.14 | 0.44 | -3.00 | 0.00 | 0.00 | 0.00 | 2.00 | ▁▁▇▁▁ |
| KIDS_3 | 0 | 1 | 0.14 | 0.42 | -1.00 | 0.00 | 0.00 | 0.00 | 2.00 | ▁▇▁▁▁ |
| KIDS_4 | 0 | 1 | 0.20 | 0.52 | -1.00 | 0.00 | 0.00 | 0.00 | 2.00 | ▁▇▁▁▁ |
| EMPLOY | 0 | 1 | 6.06 | 12.92 | 1.00 | 2.00 | 3.00 | 8.00 | 99.00 | ▇▁▁▁▁ |
| EMPLOY1_1 | 0 | 1 | 0.08 | 0.28 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| EMPLOY1_2 | 0 | 1 | 0.39 | 0.49 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▅ |
| EMPLOY1_3 | 0 | 1 | 0.09 | 0.29 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| EMPLOY1_4 | 0 | 1 | 0.06 | 0.24 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| EMPLOY1_5 | 0 | 1 | 0.04 | 0.20 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| EMPLOY1_6 | 0 | 1 | 0.05 | 0.21 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| EMPLOY1_7 | 0 | 1 | 0.04 | 0.20 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| EMPLOY1_8 | 0 | 1 | 0.30 | 0.46 | 0.00 | 0.00 | 0.00 | 1.00 | 1.00 | ▇▁▁▁▃ |
| EMPLOY1_9 | 0 | 1 | 0.02 | 0.13 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| RETIRE | 0 | 1 | -0.91 | 1.70 | -2.00 | -2.00 | -2.00 | 1.00 | 3.00 | ▇▁▁▂▁ |
| MILITARY | 0 | 1 | 0.19 | 0.41 | -1.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▁▁▇▁▂ |
| Military_Status | 0 | 1 | 4.47 | 1.19 | -1.00 | 5.00 | 5.00 | 5.00 | 5.00 | ▁▁▁▁▇ |
| agecat | 0 | 1 | 4.45 | 2.12 | 1.00 | 3.00 | 4.00 | 6.00 | 8.00 | ▇▃▇▅▆ |
| generation | 0 | 1 | 2.55 | 1.05 | 1.00 | 2.00 | 2.00 | 3.75 | 4.00 | ▃▇▁▅▆ |
| PPEDUC | 0 | 1 | 3.16 | 1.18 | 1.00 | 2.00 | 3.00 | 4.00 | 5.00 | ▂▇▇▆▅ |
| PPETHM | 0 | 1 | 1.62 | 1.08 | 1.00 | 1.00 | 1.00 | 2.00 | 4.00 | ▇▁▁▁▂ |
| PPGENDER | 0 | 1 | 1.48 | 0.50 | 1.00 | 1.00 | 1.00 | 2.00 | 2.00 | ▇▁▁▁▇ |
| PPHHSIZE | 0 | 1 | 2.52 | 1.22 | 1.00 | 2.00 | 2.00 | 3.00 | 5.00 | ▃▇▃▂▂ |
| PPINCIMP | 0 | 1 | 5.51 | 2.67 | 1.00 | 3.00 | 6.00 | 8.00 | 9.00 | ▅▅▂▆▇ |
| PPMARIT | 0 | 1 | 2.04 | 1.39 | 1.00 | 1.00 | 1.00 | 3.00 | 5.00 | ▇▁▂▂▁ |
| PPMSACAT | 0 | 1 | 0.87 | 0.34 | 0.00 | 1.00 | 1.00 | 1.00 | 1.00 | ▁▁▁▁▇ |
| PPREG4 | 0 | 1 | 2.64 | 1.03 | 1.00 | 2.00 | 3.00 | 3.00 | 4.00 | ▅▅▁▇▆ |
| PPREG9 | 0 | 1 | 5.15 | 2.53 | 1.00 | 3.00 | 5.00 | 7.00 | 9.00 | ▆▇▇▅▇ |
| PPT01 | 0 | 1 | 0.04 | 0.19 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PPT25 | 0 | 1 | 0.08 | 0.27 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PPT612 | 0 | 1 | 0.13 | 0.34 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PPT1317 | 0 | 1 | 0.12 | 0.33 | 0.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▇▁▁▁▁ |
| PPT18OV | 0 | 1 | 2.08 | 0.81 | 1.00 | 2.00 | 2.00 | 2.00 | 4.00 | ▃▇▁▂▁ |
| PCTLT200FPL | 0 | 1 | -0.08 | 1.33 | -5.00 | 0.00 | 0.00 | 0.00 | 1.00 | ▁▁▁▁▇ |
| finalwt | 0 | 1 | 1.00 | 0.59 | 0.17 | 0.60 | 0.85 | 1.25 | 6.64 | ▇▁▁▁▁ |
7.2 Appendix B: Details on Financial Well-Being Scale
According to the CFPB, “The CFPB Financial Well-Being Scale is a free tool to measure an individual’s level of financial well-being. The scale consists of 10 questions and a scoring method. The CFPB Financial Well-Being Scale can be used to 1. assess a person’s financial well-being before providing a service, 2. track changes in an individual’s financial wellbeing over time, and 3. measure the extent to which programs are improving the financial well-being of the individuals that they serve.”
The 10 questions of the scale are: - FWB1_1: I could handle a major unexpected expense - FWB1_2: I am securing my financial future - FWB1_3: Because of my money situation…I will never have the things I want in life - FWB1_4: I can enjoy life because of the way I’m managing my money - FWB1_5: I am just getting by financially - FWB1_6: I am concerned that the money I have or will save won’t last - FWB2_1: Giving a gift…would put a strain on my finances for the month - FWB2_2: I have money left over at the end of the month - FWB2_3: I am behind with my finances - FWB2_4: My finances control my life
The “CFPB Financial Well-Being Scale score is a number between 0 and 100. A higher score indicates a higher level of measured financial well-being, but there is not a specific cut-off for a “good” or “bad” financial well-being score. Most people’s scores will fall somewhere in the middle—extremely low or extremely high scores will be uncommon." Of note, scores are also adjusted for for people whose age is 62 or over, reflecting retirement age. More details can be found at: https://files.consumerfinance.gov/f/documents/201701_cfpb_FinancialWell-Being_Quick-Guide.pdf
7.3 Appendix C: Linear Regression Assumptions for Final Model
We believe the linear model assumptions are reasonably met in this case. The residuals follow a fairly symmetric pattern about h = 0 and while one could argue there is some mild grouping and funneling, they are decently evenly distributed within a band. While the ends of the qq plot could indicate some skewness for the normality assumption, it’s not particularly extreme and the line is pretty straight so we are not overly concerned about it.